Whamcloud - gitweb
yury [Fri, 13 Jun 2008 08:53:45 +0000 (08:53 +0000)]
b=15230
r=nikita,shadow
- fixed handling for OBD_FAIL_$PREF_$OPC_NET fail_ids in mdt. Former code did not
check it correctly (due to typo with && instead of &) in mdt_req_handle() and
they all did not work. In same time, some handlers like mdt_close() and
mdt_enqueue() tried to check them again (result of some wrong fix) but again, did
it not correctly. They returned 0 error without doing anything. This should
have to emulate network failure. But as they did not allocate reply buffer and
returned 0 error, they caused rs != NULL assert in ptlrpc. Fxing this also fixed
replay-single.sh test_53* and replay_dual.sh test_12 and possibly others;
- removed checking for NET fail_id in mdt_close() and mdt_enqueue() - sources
of recent assert;
- added sanity check in mdt_req_handle() for any other invalid situation about
returning 0 error and not allocating reply buffers;
- removed mdt_reply(), move its one line call into mdt_req_handle(). This was
needed to simplify handling NET fail_ids in which case we should just return 0
and make sure that no reply is sent;
- comments and cleanups;
- in reply-dual.sh - remove test 8 from ALWAYS_EXCEPT. It passes in HEAD.
Originally for placed into ALWAYS_EXCEPT for old mds code and later moved to
HEAD test scripts but as mds in HEAD is completely new this bug is making any
sense there;
- in reply-single.sh - remove tests 0b 39 56 from ALWAYS_EXCEPT. They are
passing in HEAD. Also they are obsolete and related to closed bugs.
bobijam [Fri, 13 Jun 2008 01:29:20 +0000 (01:29 +0000)]
Branch HEAD
b=14134
i=johann
minor fix handling '/etc/mtab' issue.
brian [Thu, 12 Jun 2008 20:10:18 +0000 (20:10 +0000)]
b=15870
i=adilger
Backing this out as it tries to solve a problem that doesn't really exist
and causes build breakage in doing so.
I have advised the customer that opened this bug as to how they can fix
thier OFED source pool to match the interface provided by the
kernel-ib-devel RPM, which is our reference implementation.
grev [Thu, 12 Jun 2008 08:34:41 +0000 (08:34 +0000)]
fix for
b=15711
i=Tappro
set mds${num}failover_HOST properly
fix for
b=15255
i=Adilger
umount $MOUNT2 only if it is set
bobijam [Thu, 12 Jun 2008 02:51:11 +0000 (02:51 +0000)]
Branch HEAD
b=15345
o=johann
i=nathan, bobijam
check --writeconf flag when forcing the index.
grev [Wed, 11 Jun 2008 20:11:09 +0000 (20:11 +0000)]
Cleanup wrong mount_facet message
(introduced by deen rev 1.102)
grev [Wed, 11 Jun 2008 17:43:30 +0000 (17:43 +0000)]
b=14384
i=Adilger
test_9, test_10a fixes: do not break DIR value
isaac [Wed, 11 Jun 2008 16:48:54 +0000 (16:48 +0000)]
- remove trailing spaces, expand tabs.
yangsheng [Wed, 11 Jun 2008 09:03:10 +0000 (09:03 +0000)]
Branch HEAD
b=15667
i=andrew, johann
Remove the useless function echo_client_ubrw().
shadow [Wed, 11 Jun 2008 06:48:53 +0000 (06:48 +0000)]
timeout and assert in invalidate import.
b=15716
i=rread
i=johann
adilger [Tue, 10 Jun 2008 21:43:18 +0000 (21:43 +0000)]
Branch b1_6
Ignore both Module.symvers and Modules.symvers, for SLES.
brian [Tue, 10 Jun 2008 20:10:42 +0000 (20:10 +0000)]
b=15316
i=yibin.wang
i=cliff.white:
Add the kernel-ib* RPMs to the reuse strategy.
adilger [Tue, 10 Jun 2008 17:51:07 +0000 (17:51 +0000)]
Branch b1_6
Remove obsolete tips from cvsdiffclient header. Move the "editable" comments
together, after the instructions, to be more consistent.
komaln [Tue, 10 Jun 2008 14:27:20 +0000 (14:27 +0000)]
b=14471
i=adilger
replace /proc/fs/lustre access with lctl {get,set}_param in test-framework.sh
yangsheng [Tue, 10 Jun 2008 08:04:17 +0000 (08:04 +0000)]
Branch HEAD
b=15742
i=johann, zhenyu.xu
Update to RHEL4 kernel-2.6.9-67.0.15.
bwzhou [Tue, 10 Jun 2008 07:42:25 +0000 (07:42 +0000)]
Branch HEAD
b=15759
r=rread, green
Description: MDS or OSS service threads fail startup with -24 (-EMFILE)
Details : During startup under recovery, it is possible for service
thread startup to fail in ptlrpc_start_threads() if one
of the threads begins processing a request and then starts
an additional thread. This causes ptlrpc_start_threads()
to try and start 1 too many threads and get an error.
bwzhou [Tue, 10 Jun 2008 06:59:39 +0000 (06:59 +0000)]
Branch HEAD
b=15776
r=shadow, vitaly
Prevent locks stalls on glimpse races
bwzhou [Tue, 10 Jun 2008 06:26:59 +0000 (06:26 +0000)]
Branch HEAD
b=13380
remove test 2 and 4 form ALWAYS_EXCEPT since 13380(13129) is fixed
bwzhou [Tue, 10 Jun 2008 05:50:35 +0000 (05:50 +0000)]
Branch HEAD
b=15436
r=nathan, bobijam
remove failover mgs setting from mkfs_lustre
yangsheng [Tue, 10 Jun 2008 04:22:50 +0000 (04:22 +0000)]
Branch HEAD
b=15592
i=oleg, shadow
Correctly handle the FS_RENAME_DOES_D_MOVE flag in Lustre.
robert.read [Mon, 9 Jun 2008 22:58:36 +0000 (22:58 +0000)]
Branch HEAD
b=15929
i=grev
Cleanup PATH in tests so absolute paths are used, and
each path is added at most once.
grev [Mon, 9 Jun 2008 20:22:28 +0000 (20:22 +0000)]
b=16019
i=Maxim.Patlasov
get rid of quotation in "options lnet"
grev [Fri, 6 Jun 2008 16:37:35 +0000 (16:37 +0000)]
b=15979
i=Brian
i=Alexey.Lyashkov
Set build_success=true if build_lustre did not fail for case ! PATCHLESS and ! -z LINUX
huanghua [Fri, 6 Jun 2008 02:19:20 +0000 (02:19 +0000)]
Branch HEAD
b=15552
i=nikita.danilov
i=adilger
- Igif support for the case when of HEAD based server runs on old disk format
robert.read [Thu, 5 Jun 2008 23:11:56 +0000 (23:11 +0000)]
Branch HEAD
b=13537
i=grev
The old test was using the used space rather than free space. This fixes
that and also cuts the amount by half, so we don't completely fill up the
fs. It also caps the size at 1GB, like the previous version.
robert.read [Thu, 5 Jun 2008 23:08:22 +0000 (23:08 +0000)]
Branch HEAD
b=15928
i=grev
Reset the current directory after the each test completes.
deen [Thu, 5 Jun 2008 18:39:58 +0000 (18:39 +0000)]
Add missed files from previous commit.
b=10718
grev [Thu, 5 Jun 2008 16:06:27 +0000 (16:06 +0000)]
b=14949
i=Adilger
integrate lustre-iokit into normal build process
deen [Thu, 5 Jun 2008 11:43:21 +0000 (11:43 +0000)]
Directly associate cached pages to lock that protect those pages,
this allows us to quickly find what pages to write and remove
once lock callback is received.
b=10718
i=green
i=johann
isaac [Thu, 5 Jun 2008 03:37:22 +0000 (03:37 +0000)]
i=liang:
- remove liblustre.h dependency from selftest.
isaac [Thu, 5 Jun 2008 03:33:40 +0000 (03:33 +0000)]
i=liang:
- turn SRPC_PEER_CREDITS into a module parameter, which is useful for
high-latency networks.
deen [Wed, 4 Jun 2008 17:56:05 +0000 (17:56 +0000)]
Change LASSERTs to client eviction (i.e. abort client's recovery)
because LASSERT on both the data supplied by a client, and the data
on disk is dangerous and incorrect.
b=15574
i=adilger
i=shadow
fanyong [Wed, 4 Jun 2008 15:42:33 +0000 (15:42 +0000)]
Merge b_head_capa from HEAD (20080604_1758)
b=15563
yangsheng [Wed, 4 Jun 2008 14:38:10 +0000 (14:38 +0000)]
Branch HEAD
b=15741
i=johann, zhenyu.xu
Update kernel to RHEL5 kernel-2.6.18-53.1.19.el5.
yury [Wed, 4 Jun 2008 12:07:32 +0000 (12:07 +0000)]
b=15226
r=shadow,nikita
- fixes issue with accessing freed @ns on behalf rpc handling when some late rpc comes in fini time;
- add new comments in functions, converts existing ones to be inline wigth doxygen comments policies.
girish [Tue, 3 Jun 2008 11:26:07 +0000 (11:26 +0000)]
Add block map validation
b=13144
i=adilger
i=kalpak
yury [Tue, 3 Jun 2008 08:04:50 +0000 (08:04 +0000)]
b=15440
- missed part of 15440. Do not forget to call ldlm_pool_fini() before removing ns proc dir;
- call ldlm_namespace_unregister() namespace _before_ __ldlm_namespace_free() which is waiting for refc == 0 as ldlm_namespace_unregister() drops refc.
komaln [Tue, 3 Jun 2008 06:11:51 +0000 (06:11 +0000)]
b=14471
r=nathan, andreas
To add -N option in lctl get_param which will print only path names.
grev [Mon, 2 Jun 2008 10:19:01 +0000 (10:19 +0000)]
b=15711
i=Tappro
remove obsolete insanity-local cfg
grev [Mon, 2 Jun 2008 09:29:56 +0000 (09:29 +0000)]
b=15711
i=Tappro
-- t-f and insanity fix to work properly with FAILURE_MODE=HARD
-- remove insanity-local cfg
nikita [Sun, 1 Jun 2008 21:07:04 +0000 (21:07 +0000)]
update to doxygen templates
yury [Fri, 30 May 2008 09:24:33 +0000 (09:24 +0000)]
b=15440
r=shadow, nikita
- fixes missed list init in CFS_LIST_INIT_HEAD(&ns->ns_list_chain);
- cleanups about registering/unregistering @ns in its list.
yangsheng [Fri, 30 May 2008 08:17:04 +0000 (08:17 +0000)]
Branch HEAD
b=15594
i=adilger, shadow
The dd has a change of behaviour in 2.6.22 kernel. It might return
a partial read in read() system-call in some case.
stinson1947 [Fri, 30 May 2008 05:23:29 +0000 (05:23 +0000)]
Made changes to allow b1_8 to get it's components from the right
repositories.
bobijam [Fri, 30 May 2008 02:56:38 +0000 (02:56 +0000)]
Branch HEAD
b=14134
i=nathan, johann
Description: enable MGS and MDT services start separately
Details : add a 'nomgs' option in mount.lustre to enable start a MDT with
a co-located MGS without starting the MGS, which is a complement
to 'nosvc' mount option.
vitaly [Thu, 29 May 2008 21:46:00 +0000 (21:46 +0000)]
Branch HEAD
b=15904
i=wangdi
i=shadow
clean log_commit_master properly at cleanup in llog_commit_thread()
bobijam [Thu, 29 May 2008 07:24:38 +0000 (07:24 +0000)]
restore 51d to ALWAYS_EXCEPT list. last commit accidentally removed it.
pravins [Thu, 29 May 2008 06:35:13 +0000 (06:35 +0000)]
update module tags for new uMDS branch.
bobijam [Thu, 29 May 2008 02:52:13 +0000 (02:52 +0000)]
Branch HEAD
b=13310
o=eeb
i=joahnn, nikita
The bug is that target_send_reply_msg() calls ptlrpc_error() if 'rc != 0', but
"difficult" replies may not be passed to rpc_error().
vs [Wed, 28 May 2008 23:49:03 +0000 (23:49 +0000)]
b=15346
i=green,shadow,vitaly
skiplists are used to group compatible locks on granted list
that was implemented as tracking first and last lock of each lock group
the patch changes that to using doubly linked circular lists
adilger [Wed, 28 May 2008 23:39:49 +0000 (23:39 +0000)]
Branch HEAD
Add updated racer.sh script
i=oleg.drokin
grev [Wed, 28 May 2008 20:34:07 +0000 (20:34 +0000)]
b=12622
add 51d to ALWAYS_EXCEPT list
grev [Wed, 28 May 2008 20:07:21 +0000 (20:07 +0000)]
b=15876
i=Adilger
skip LFSCK test for remote MDS also
grev [Wed, 28 May 2008 19:49:15 +0000 (19:49 +0000)]
b=15906
i=Eric.Mei
mark SANITY_SEC as "Done" when completed
grev [Wed, 28 May 2008 19:37:13 +0000 (19:37 +0000)]
b=14773
i=Adilger
rename subtests to skipping mechanism works properly
brian [Wed, 28 May 2008 19:17:16 +0000 (19:17 +0000)]
b=15908
i=johann
i=zhenyu.xu:
Small message correction.
brian [Wed, 28 May 2008 18:38:19 +0000 (18:38 +0000)]
b=15482
i=cliffw
i=wangyb
A forgotten small portion of the previous commit for this bug.
komaln [Wed, 28 May 2008 06:04:19 +0000 (06:04 +0000)]
b=14471
r=adilger
to replace the sysctl lustre.* and lnet.* instances with lctl {get,set}_param
girish [Tue, 27 May 2008 17:48:51 +0000 (17:48 +0000)]
Upstream fix memory leak when verifying checksums in the journal
nikita [Tue, 27 May 2008 16:18:44 +0000 (16:18 +0000)]
interface documentation: add doxygen templates and publish script.
adilger [Mon, 26 May 2008 20:06:39 +0000 (20:06 +0000)]
Branch HEAD
Add clarifying comment.
adilger [Mon, 26 May 2008 19:53:14 +0000 (19:53 +0000)]
Branch HEAD
Correct comment.
adilger [Mon, 26 May 2008 19:52:11 +0000 (19:52 +0000)]
Branch HEAD
Use 64-bit XID field for async commit errors.
b=2066
i=emoly.liu
i=nathan.rutman
adilger [Mon, 26 May 2008 19:47:14 +0000 (19:47 +0000)]
Branch HEAD
Elaborate on why the _ASM_* definitions are set.
b=14746
adilger [Mon, 26 May 2008 19:46:10 +0000 (19:46 +0000)]
Branch HEAD
Fix compile warning for printf.
adilger [Mon, 26 May 2008 19:44:38 +0000 (19:44 +0000)]
Branch HEAD
Update llbackup.8 man page to match b1_6.
b=14711
adilger [Mon, 26 May 2008 19:42:28 +0000 (19:42 +0000)]
Branch HEAD
Add "dk" to common lctl commands.
yangsheng [Mon, 26 May 2008 13:49:46 +0000 (13:49 +0000)]
Branch HEAD
b=15649
i=oleg, shadow
Enable debug options in vanilla kernel config file.
grev [Mon, 26 May 2008 10:43:03 +0000 (10:43 +0000)]
b=15874
i=Adilger
skip test_32a, test_32b for NETTYPE != tcp
anserper [Sun, 25 May 2008 19:44:00 +0000 (19:44 +0000)]
Branch HEAD
b=14322
adilger [Sun, 25 May 2008 05:45:26 +0000 (05:45 +0000)]
Branch HEAD
Update which_patch to indicate support for 2.6.22 patchless.
bwzhou [Sat, 24 May 2008 13:27:52 +0000 (13:27 +0000)]
Branch HEAD
b=11777
r=green, shadow
invalidate the import to halt request handling so the namespace can be safely
cleaned up
girish [Sat, 24 May 2008 10:57:18 +0000 (10:57 +0000)]
Update e2fsprogs version to 1.40.7-sun3
shadow [Fri, 23 May 2008 15:57:10 +0000 (15:57 +0000)]
fix build lustre with o2ib and SLES10.
b=15870
i=walter
shadow [Fri, 23 May 2008 15:51:55 +0000 (15:51 +0000)]
fix build lustre with o2ib and SLES10.
b=15870
i=walter
komaln [Fri, 23 May 2008 07:19:05 +0000 (07:19 +0000)]
b=14471
r=Adilger
to remove instances of /proc/sys in sanity.sh by using either sysctl or lctl {get,set}_param
shadow [Fri, 23 May 2008 06:02:49 +0000 (06:02 +0000)]
don't create too many objects after del orphan.
b=14835
i=tappro
i=deen
grev [Thu, 22 May 2008 18:45:36 +0000 (18:45 +0000)]
b=15255
i=Adilger
config file for several remote clients; replay-single test_70a, test_70b
girish [Thu, 22 May 2008 17:01:56 +0000 (17:01 +0000)]
Fix ext3 inode version field being filled with 0x5a5a5a5a on disk
b=15604
i=adilger
kalpak [Thu, 22 May 2008 16:38:19 +0000 (16:38 +0000)]
b=9570
i=adilger, girish
Print dir inode number when displaying corrupt hash code warning
grev [Thu, 22 May 2008 12:27:57 +0000 (12:27 +0000)]
revert my previous commit due to it brokes rpm build
grev [Thu, 22 May 2008 10:09:14 +0000 (10:09 +0000)]
b=14949
i=Adilger
integrate lustre-iokit into normal build process
shadow [Thu, 22 May 2008 05:14:12 +0000 (05:14 +0000)]
fix build on rhel5/ia64.
b=15851
i=zam
robert.read [Wed, 21 May 2008 17:52:46 +0000 (17:52 +0000)]
Branch HEAD
b=13285
Reverting the rest of the script libexecdir patch to fix rpm builds.
fanyong [Wed, 21 May 2008 14:20:44 +0000 (14:20 +0000)]
Land b1_8_dir_ra onto HEAD (20080521_1834)
b=11401,15405
i=huanghua
i=tappro
adilger [Wed, 21 May 2008 04:44:58 +0000 (04:44 +0000)]
Branch HEAD
Quiet build warnings when enabling MPI test programs.
robert.read [Wed, 21 May 2008 00:55:39 +0000 (00:55 +0000)]
Branch b_release_1_6_5
b=13285
Perhaps the libexecdir scripts patch was only partially applied or reverted. This
patch reverts the change apparently made to lustre.spec.in back to libdir. So
Terry can build the next RC.
bwzhou [Tue, 20 May 2008 15:09:52 +0000 (15:09 +0000)]
Branch HEAD
b=13285
r=green, bobijam
use the libexec convention for scripts
bwzhou [Tue, 20 May 2008 15:07:33 +0000 (15:07 +0000)]
Branch b1_6
b=13285
r=green, bobijam
use the libexec convention for scripts
bwzhou [Tue, 20 May 2008 13:02:22 +0000 (13:02 +0000)]
Branch HEAD
b=13285
r=adilger, deen
1. updated ext3-mballoc3-core.patch and added ext3-max-dir-size.patch to limit the maximum size of a directory;
2. updated other ldiskfs patches to resolve patch conflicts caused by the above two new patches;
3. added a test case in sanity.sh.
bwzhou [Tue, 20 May 2008 09:07:37 +0000 (09:07 +0000)]
Branch b1_6
b=13285
r=adilger, deen
1. updated ext3-mballoc3-core.patch and added ext3-max-dir-size.patch to limit the maximum size of a directory;
2. updated other ldiskfs patches to resolve patch conflicts caused by the above two new patches;
3. added a test case in sanity.sh.
komaln [Tue, 20 May 2008 08:43:34 +0000 (08:43 +0000)]
b=14471
r=Andreas
replace /proc/fs/lustre access with lctl set_param
shadow [Tue, 20 May 2008 06:44:43 +0000 (06:44 +0000)]
fix ASSERTION(client_stat->nid_exp_ref_count == 0) during
run acc-small.
b=15139
i=tappro
i=komal
fanyong [Tue, 20 May 2008 01:59:10 +0000 (01:59 +0000)]
Branch HEAD
b=15828
i=adilger
i=umka
return "-EIO" when write to read only proc interface which "write_proc" is NULL.
bobijam [Tue, 20 May 2008 01:57:51 +0000 (01:57 +0000)]
Branch HEAD
b=15575
i=wangdi, johann, green, shadow
Description: Stack overflow during MDS log replay
Details : ease stack pressure by using a thread dealing llog_process.
bobijam [Tue, 20 May 2008 01:26:32 +0000 (01:26 +0000)]
Branch b1_6
b=15575
i=wangdi, johann, green, shadow
Description: Stack overflow during MDS log replay
Details : ease stack pressure by using a thread dealing llog_process.
shadow [Mon, 19 May 2008 16:21:38 +0000 (16:21 +0000)]
fix typo which added with 2.6.22 support.
pravins [Mon, 19 May 2008 15:27:17 +0000 (15:27 +0000)]
b=14230
i=amit.sharma
i=manoj.joseph
umds cleanup: use libcfs kmap api rather than kernel kmap apis.
pravins [Mon, 19 May 2008 15:24:49 +0000 (15:24 +0000)]
b=14230
i=rahul.deshmukh
i=manoj.joseph
umds cleanup: use libcfs api.
pravins [Mon, 19 May 2008 15:23:10 +0000 (15:23 +0000)]
b=14230
i=amit.sharma
i=tappro
umds cleanup: use pid instead of current * to have portability.